Retention Time Prediction in High-Performance Liquid Chromatography Using Random Forest Regression

Retention Time Prediction in High-Performance Liquid Chromatography Using Random Forest Regression | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Retention Time Prediction in High-Performance Liquid Chromatography Using Random Forest Regression Varshini Ganesan Selvi, Anirudh R Urs, Trilok Chandran B This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6053718/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract High-Performance Liquid Chromatography is widely used for compound analysis, where retention time (RT) serves as a critical parameter. This study employs a Random Forest Regression model to predict RT based on molecular descriptors such as molecular weight, partial charge, partition coefficient, and topological polar surface area. The model successfully predicted the retention time with high similarity to the real data, thereby validating its accuracy. This study highlights the potential of machine learning in optimizing chromatographic analysis. High-Performance Liquid Chromatography retention time Random Forest Regression molecular descriptors Figures Figure 1 Figure 2 Figure 3 Figure 4 I. INTRODUCTION High-Performance Liquid Chromatography (HPLC) is a widely used analytical technique in chemistry, biochemistry, and pharmaceutical sciences for separating, identifying, and quantifying components in complex mixtures. It operates by passing a liquid sample through a column packed with stationary phase particles while being propelled by a high-pressure mobile phase. The interaction between the analyte, mobile phase, and stationary phase dictates the separation process, leading to different retention times (RTs) for various compounds. Due to its high resolution, sensitivity, and versatility, HPLC is extensively used in drug discovery, environmental analysis, and biomolecular research [1–3]. Retention time (RT) is a crucial parameter in HPLC that refers to the time a compound takes to pass through the column from injection to detection. It serves as an essential characteristic for compound identification and plays a critical role in method development and optimization. RT is influenced by various physicochemical properties of the analyte, including molecular weight, polarity, and interactions with the stationary and mobile phases. Understanding RT behaviour is vital for ensuring accurate separations, reproducibility, and efficient method development [4–5]. Several molecular descriptors influence the retention time in HPLC [6–7]. Molecular weight (Mol Wt) affects diffusion rates and interaction strengths within the column, with larger molecules typically exhibiting longer retention due to steric hindrance. Topological Polar Surface Area (TPSA) correlates with hydrogen bonding potential, where highly polar compounds with large TPSA values tend to elute faster in reversed-phase HPLC. LogP (partition coefficient) represents the compound’s lipophilicity, influencing its affinity towards the stationary phase—higher LogP values often result in longer RTs in non-polar stationary phases. Partial charge distribution impacts solubility and interactions within the chromatographic system, affecting peak shapes and separation efficiency [8–9]. By understanding these molecular descriptors, RT predictions become more accurate, aiding in chromatographic method development [10–12]. Predicting retention time in advance of performing HPLC experiments offers several advantages. It significantly reduces the trial-and-error approach in method development, saving both time and resources. A reliable RT prediction model enhances compound identification, particularly in metabolomics and pharmaceutical research, where large datasets require efficient analysis. Moreover, predictive models aid in optimizing column conditions and mobile phase compositions, ensuring better resolution and peak efficiency. This capability is particularly valuable for high-throughput analytical workflows, where precise retention time estimation streamlines the experimental process [13–15]. However, certain factors affect the accuracy and generalizability of RT prediction models. Retention time is highly dependent on column selection, mobile phase composition, and gradient conditions, meaning that models trained under specific conditions may not generalize well if these parameters are altered [16]. Additionally, matrix effects, such as interactions with contaminants or co-eluting compounds, can cause deviations between predicted and actual RT values. While no model can fully replace direct experimental validation, a well-trained RT prediction model can serve as a valuable pre-screening tool [17]. By providing an estimate of chromatographic behaviour, it enables researchers to optimize experimental conditions and reduce unnecessary resource expenditure before conducting expensive HPLC-MS analyses [18–21]. Machine learning approaches have recently gained traction in retention time prediction, with Random Forest Regression emerging as a robust and interpretable model. Random Forest Regression is an ensemble learning technique that constructs multiple decision trees to predict continuous variables, such as RT, by averaging outputs from individual trees. This method effectively captures complex nonlinear relationships between molecular descriptors and retention times, offering high predictive accuracy. Additionally, Random Forest models handle large datasets efficiently, are resistant to overfitting, and provide insights into the importance of various molecular features in RT prediction [13,22]. This research aims to develop and validate a Random Forest Regression-based model for predicting retention times in HPLC. By integrating molecular descriptors as input features, we seek to enhance chromatographic efficiency, reduce experimental workload, and contribute to advancing computational approaches in analytical chemistry [23–27]. The findings of this study hold the potential to improve HPLC method development and facilitate data-driven optimization in separation sciences. II. METHODOLOGY The dataset used in this study was obtained from Kaggle and contained molecular descriptors along with retention time values derived from high-performance liquid chromatography (HPLC) experiments [28]. The dataset included key molecular features such as the logarithm of the partition coefficient (MolLogP), molecular weight (MolWt), topological polar surface area (TPSA), and partial charge, which were considered as independent variables for predictive modeling. Before model training, the dataset underwent preprocessing to ensure data integrity and consistency. Any missing values were removed to maintain the quality of the input data. Additionally, the independent variables were standardized to eliminate scale differences and ensure uniform contribution to the model’s predictions. The processed dataset was then split into training and testing subsets, with 80% allocated for training and 20% reserved for model evaluation. For predictive modeling, a Random Forest Regressor (RFR) was implemented due to its capability to handle complex, non-linear relationships within the data. The number of decision trees (ntree) was set to 50, striking a balance between computational efficiency and predictive performance. The model was trained on the selected independent variables using the training dataset and subsequently tested on the unseen test set to evaluate its generalization ability. The entire implementation was executed in R programming language within the RStudio Integrated Development Environment (IDE). The code used for data preprocessing, model training, and performance evaluation is included below. III. RESULTS The performance of the trained model was evaluated using standard regression metrics, which provide a comprehensive assessment of its predictive accuracy. The results, presented in the table below, highlight the model’s ability to estimate retention time effectively based on the selected molecular descriptors. Table 1. Performance metrics table S.No. Performance Metrics Value 1 Mean Squared Error 109.84 2 R-squared 0.609 3 Mean Absolute Error 80.02 A scatter plot comparing actual vs. predicted retention times showed a dense clustering of points around the ideal prediction line. This distribution indicates that the model successfully captures the underlying relationship between retention time and molecular properties, demonstrating its reliability in making accurate predictions IV. CONCLUSION This study demonstrates the effectiveness of Random Forest Regression in predicting HPLC retention time based on molecular descriptors. The model achieves an R² value of 0.609, indicating a moderate predictive capability. Future work can improve accuracy by incorporating additional molecular descriptors and further optimizing hyperparameters, paving the way for integration of machine learning in chromatography analysis. Declarations Author Contribution V.G.S , A.R.U & T.C.B wrote and edited the main manuscript text. All authors reviewed the manuscript. Acknowledgement We express our sincere gratitude to RV College of Engineering and RV Sikshana Samithi Trust. Data Availability The real time data set was obtained from : Kaggle, “Meltin retention times with molecular descriptors.” [Online]. Available: https://www.kaggle.com/datasets/satwikmurarka/meltin-retention-times-with-molecular-descriptors. References P. Patel, S. B. Narkhede, and S. Luhar, “High-performance liquid chromatography (HPLC): A comprehensive review,” EPRA Int. J. Res. Dev., pp. 85–86, 2024. R. Ahmed, “High-Performance Liquid Chromatography (HPLC): Principles, applications, versatility, efficiency, innovation and comparative analysis in modern analytical chemistry and in pharmaceutical sciences,” 2024. H. A. Gavit, A. R. Pawar, V. V. Patil, P. S. Patil, and P. H. Sisodiya, “Advancement in High-Performance Liquid Chromatography Techniques: A comprehensive review,” Asian J. Pharm. Anal., vol. 95, pp. 95–103, 2024, doi: 10.52711/2231-5675.2024.00017 . “Characterization of a high throughput approach for large scale retention measurement in liquid chromatography,” 2023, doi: 10.26434/chemrxiv-2023-czt1p-v3 . “RT-Tranformer: Retention time prediction for metabolite annotation to assist in metabolite identification,” 2023, doi: 10.26434/chemrxiv-2023-pf268 . J. L. E. Reubsaet and K. Jinno, “Characterisation of important interactions controlling retention behaviour of analytes in reversed-phase high-performance liquid chromatography,” TrAC Trends Anal. Chem., vol. 17, no. 3, pp. 157–166, 1998, doi: 10.1016/S0165-9936(97)00127-1 . J. Golubović, A. Protić, B. Otašević, and M. Zečević, “Quantitative structure-retention relationships applied to development of liquid chromatography gradient-elution method for the separation of sartans,” Talanta, vol. 150, pp. 190–197, 2016, doi: 10.1016/j.talanta.2015.12.035 . V. David and A. Medvedovici, “Structure-retention correlation in liquid chromatography for pharmaceutical applications,” J. Liq. Chromatogr. Relat. Technol., vol. 30, pp. 761–789, 2007, doi: 10.1080/10826070701191052 . T. Bączek, R. Kaliszan, K. Novotná, and P. Jandera, “Comparative characteristics of HPLC columns based on quantitative structure-retention relationships (QSRR) and hydrophobic-subtraction model,” J. Chromatogr. A, vol. 1075, no. 1–2, pp. 109–115, 2005, doi: 10.1016/J.CHROMA.2005.03.117 . P. P. Basuri, K. Sahini, N. C. N, and G. V., “Retention behaviour of analytes in reversed phase high performance liquid chromatography - A review,” Biomed. Chromatogr., vol. 37, no. 7, p. e5482, 2022, doi: 10.1002/bmc.5482 . Y. Wang, Retention Studies in HPLC , 1992. U. Judycka, K. Jagiello, M. Gromelski, L. Bober, J. Błażejowski, and T. Puzyn, “Chemometric outlook on correlations between retention parameters of polar and semipolar HPLC columns and physicochemical characteristics of ampholytic substances of biological and pharmaceutical relevance,” Struct. Chem., vol. 29, no. 6, pp. 1839–1844, 2018, doi: 10.1007/S11224-018-1174-5 . S. Osipenko et al., “Machine learning to predict retention time of small molecules in nano-HPLC,” Anal. Bioanal. Chem., vol. 412, pp. 7767–7776, 2020, doi: 10.1007/s00216-020-02905-0 . L. Sun et al., “A simple method for HPLC retention time prediction: Linear calibration using two reference substances,” Chin. Med., vol. 12, 2017, doi: 10.1186/s13020-017-0137-x . S. Osipenko, E. Nikolaev, and Y. Kostyukevich, “Retention time prediction with message-passing neural networks,” Separations, vol. 9, no. 10, p. 291, 2022, doi: 10.3390/separations9100291 . M. Rosés, I. Canals, H. Allemann, K. Siigur, and E. Bosch, “Retention of ionizable compounds on HPLC. 2. Effect of pH, ionic strength, and mobile phase composition on the retention of weak acids,” Anal. Chem., vol. 68, no. 23, pp. 4094–4100, 1996, doi: 10.1021/ac960105d . R. Ju et al., “Deep neural network pretrained by weighted autoencoders and transfer learning for retention time prediction of small molecules,” Anal. Chem., vol. 93, no. 47, pp. 15651–15658, 2021, doi: 10.1021/ACS.ANALCHEM.1C03250 . C. Ren et al., “Improved peptide retention time prediction in liquid chromatography through deep learning,” Anal. Chem., vol. 90, no. 18, pp. 10881–10888, 2018, doi: 10.1021/acs.analchem.8b02386 . R. Bouwmeester, L. Martens, and S. Degroeve, “Comprehensive and empirical evaluation of machine learning algorithms for small molecule LC retention time prediction,” Anal. Chem., vol. 91, no. 5, pp. 3694–3703, 2019, doi: 10.1021/acs.analchem.8b05820 . Q. Yang, H. Ji, H. Lu, and Z. Zhang, “Prediction of liquid chromatographic retention time with graph neural networks to assist in small molecule identification,” Anal. Chem., 2021, doi: 10.1021/acs.analchem.0c04071 . Q. Yang, H. Ji, X. Fan, Z. Zhang, and H. Lu, “Retention time prediction in hydrophilic interaction liquid chromatography with graph neural network and transfer learning,” J. Chromatogr. A, vol. 1656, p. 462536, 2021, doi: 10.1016/j.chroma.2021.462536 . GeeksforGeeks, “Random forest regression in Python.” [Online]. Available: https://www.geeksforgeeks.org/random-forest-regression-in-python/ . N. Goudarzi, D. Shahsavani, F. Emadi-Gandaghi, and M. Chamjangali, “Quantitative structure-property relationships of retention indices of some sulfur organic compounds using random forest technique as a variable selection and modeling method,” J. Sep. Sci., vol. 39, no. 19, pp. 3835–3842, 2016, doi: 10.1002/jssc.201600358 . N. Goudarzi, D. Shahsavani, F. Emadi-Gandaghi, and M. Chamjangali, “Application of random forests method to predict the retention indices of some polycyclic aromatic hydrocarbons,” J. Chromatogr. A, vol. 1333, pp. 25–31, 2014, doi: 10.1016/j.chroma.2014.01.048 . N. Goudarzi and D. Shahsavani, “Application of a random forests (RF) method as a new approach for variable selection and modelling in a QSRR study to predict the relative retention time of some polybrominated diphenylethers (PBDEs),” Anal. Methods, vol. 4, pp. 3733–3738, 2012, doi: 10.1039/C2AY25484K . M. Kursa, L. Komsta, and W. Rudnicki, “Random forest models of the retention constants in thin layer chromatography,” arXiv preprint, arXiv:1106.3361, 2011. E. Bandini et al., “Physicochemical modelling of retention mechanisms using machine learning,” J. Cheminform., vol. 16, p. 72, 2024. Kaggle, “Meltin retention times with molecular descriptors.” [Online]. Available: https://www.kaggle.com/datasets/satwikmurarka/meltin-retention-times-with-molecular-descriptors . Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6053718","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":421862037,"identity":"9bead87f-82d5-4b50-9f28-9a43fed2a2d7","order_by":0,"name":"Varshini Ganesan Selvi","email":"data:image/png;base64,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","orcid":"","institution":"RV College of Engineering","correspondingAuthor":true,"prefix":"","firstName":"Varshini","middleName":"Ganesan","lastName":"Selvi","suffix":""},{"id":421862038,"identity":"7f8351fa-bf4f-437c-b5dc-3da6aab1ddd3","order_by":1,"name":"Anirudh R Urs","email":"","orcid":"","institution":"RV College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Anirudh","middleName":"R","lastName":"Urs","suffix":""},{"id":421862039,"identity":"13d8f248-460a-4a4c-9da3-05a8a0505567","order_by":2,"name":"Trilok Chandran B","email":"","orcid":"","institution":"RV College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Trilok","middleName":"Chandran","lastName":"B","suffix":""}],"badges":[],"createdAt":"2025-02-18 07:38:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6053718/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6053718/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":77749095,"identity":"ab815a76-6143-45f3-9566-197335c9d6e4","added_by":"auto","created_at":"2025-03-05 07:18:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":42369,"visible":true,"origin":"","legend":"\u003cp\u003eMethodology\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6053718/v1/df828f338d959a8b70bfbc7b.png"},{"id":77748992,"identity":"2d7b03e2-73c9-44d4-8f0f-5726c185b532","added_by":"auto","created_at":"2025-03-05 07:10:47","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":145476,"visible":true,"origin":"","legend":"\u003cp\u003eHPLC Molecular Descriptor Dataset from Kaggle\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6053718/v1/01ca4b07a3ed86d8e4bc6c5c.jpeg"},{"id":77748976,"identity":"3bf96b5d-5c11-49c2-8149-58bfb52121eb","added_by":"auto","created_at":"2025-03-05 07:10:46","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":248441,"visible":true,"origin":"","legend":"\u003cp\u003eRandom Forest Regression Code Snippet\u003c/p\u003e","description":"","filename":"floatimage33.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6053718/v1/f7a6038a4b76bbd61469dfda.jpeg"},{"id":77748988,"identity":"e2553872-3548-4f68-bdb4-fd0faeee6783","added_by":"auto","created_at":"2025-03-05 07:10:47","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":10987,"visible":true,"origin":"","legend":"\u003cp\u003eScatter Plot of Actual vs Predicted RT\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6053718/v1/e56be96e5302b5ff498be51c.png"},{"id":77750357,"identity":"b5218644-f632-4f89-b6ac-5dcaac75c6a7","added_by":"auto","created_at":"2025-03-05 07:26:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":763243,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6053718/v1/78ca658a-91f5-4e77-b5f8-84677fa04a2c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Retention Time Prediction in High-Performance Liquid Chromatography Using Random Forest Regression","fulltext":[{"header":"I.\tINTRODUCTION","content":"\u003cp\u003eHigh-Performance Liquid Chromatography (HPLC) is a widely used analytical technique in chemistry, biochemistry, and pharmaceutical sciences for separating, identifying, and quantifying components in complex mixtures. It operates by passing a liquid sample through a column packed with stationary phase particles while being propelled by a high-pressure mobile phase. The interaction between the analyte, mobile phase, and stationary phase dictates the separation process, leading to different retention times (RTs) for various compounds. Due to its high resolution, sensitivity, and versatility, HPLC is extensively used in drug discovery, environmental analysis, and biomolecular research [1\u0026ndash;3].\u003c/p\u003e \u003cp\u003eRetention time (RT) is a crucial parameter in HPLC that refers to the time a compound takes to pass through the column from injection to detection. It serves as an essential characteristic for compound identification and plays a critical role in method development and optimization. RT is influenced by various physicochemical properties of the analyte, including molecular weight, polarity, and interactions with the stationary and mobile phases. Understanding RT behaviour is vital for ensuring accurate separations, reproducibility, and efficient method development [4\u0026ndash;5].\u003c/p\u003e \u003cp\u003eSeveral molecular descriptors influence the retention time in HPLC [6\u0026ndash;7]. Molecular weight (Mol Wt) affects diffusion rates and interaction strengths within the column, with larger molecules typically exhibiting longer retention due to steric hindrance. Topological Polar Surface Area (TPSA) correlates with hydrogen bonding potential, where highly polar compounds with large TPSA values tend to elute faster in reversed-phase HPLC. LogP (partition coefficient) represents the compound\u0026rsquo;s lipophilicity, influencing its affinity towards the stationary phase\u0026mdash;higher LogP values often result in longer RTs in non-polar stationary phases. Partial charge distribution impacts solubility and interactions within the chromatographic system, affecting peak shapes and separation efficiency [8\u0026ndash;9]. By understanding these molecular descriptors, RT predictions become more accurate, aiding in chromatographic method development [10\u0026ndash;12].\u003c/p\u003e \u003cp\u003ePredicting retention time in advance of performing HPLC experiments offers several advantages. It significantly reduces the trial-and-error approach in method development, saving both time and resources. A reliable RT prediction model enhances compound identification, particularly in metabolomics and pharmaceutical research, where large datasets require efficient analysis. Moreover, predictive models aid in optimizing column conditions and mobile phase compositions, ensuring better resolution and peak efficiency. This capability is particularly valuable for high-throughput analytical workflows, where precise retention time estimation streamlines the experimental process [13\u0026ndash;15].\u003c/p\u003e \u003cp\u003eHowever, certain factors affect the accuracy and generalizability of RT prediction models. Retention time is highly dependent on column selection, mobile phase composition, and gradient conditions, meaning that models trained under specific conditions may not generalize well if these parameters are altered [16]. Additionally, matrix effects, such as interactions with contaminants or co-eluting compounds, can cause deviations between predicted and actual RT values. While no model can fully replace direct experimental validation, a well-trained RT prediction model can serve as a valuable pre-screening tool [17]. By providing an estimate of chromatographic behaviour, it enables researchers to optimize experimental conditions and reduce unnecessary resource expenditure before conducting expensive HPLC-MS analyses [18\u0026ndash;21].\u003c/p\u003e \u003cp\u003eMachine learning approaches have recently gained traction in retention time prediction, with Random Forest Regression emerging as a robust and interpretable model. Random Forest Regression is an ensemble learning technique that constructs multiple decision trees to predict continuous variables, such as RT, by averaging outputs from individual trees. This method effectively captures complex nonlinear relationships between molecular descriptors and retention times, offering high predictive accuracy. Additionally, Random Forest models handle large datasets efficiently, are resistant to overfitting, and provide insights into the importance of various molecular features in RT prediction [13,22].\u003c/p\u003e \u003cp\u003eThis research aims to develop and validate a Random Forest Regression-based model for predicting retention times in HPLC. By integrating molecular descriptors as input features, we seek to enhance chromatographic efficiency, reduce experimental workload, and contribute to advancing computational approaches in analytical chemistry [23\u0026ndash;27]. The findings of this study hold the potential to improve HPLC method development and facilitate data-driven optimization in separation sciences.\u003c/p\u003e"},{"header":"II.\tMETHODOLOGY","content":" \u003cp\u003eThe dataset used in this study was obtained from Kaggle and contained molecular descriptors along with retention time values derived from high-performance liquid chromatography (HPLC) experiments [28]. The dataset included key molecular features such as the logarithm of the partition coefficient (MolLogP), molecular weight (MolWt), topological polar surface area (TPSA), and partial charge, which were considered as independent variables for predictive modeling.\u003c/p\u003e \u003cp\u003eBefore model training, the dataset underwent preprocessing to ensure data integrity and consistency. Any missing values were removed to maintain the quality of the input data. Additionally, the independent variables were standardized to eliminate scale differences and ensure uniform contribution to the model\u0026rsquo;s predictions. The processed dataset was then split into training and testing subsets, with 80% allocated for training and 20% reserved for model evaluation.\u003c/p\u003e \u003cp\u003eFor predictive modeling, a Random Forest Regressor (RFR) was implemented due to its capability to handle complex, non-linear relationships within the data. The number of decision trees (ntree) was set to 50, striking a balance between computational efficiency and predictive performance. The model was trained on the selected independent variables using the training dataset and subsequently tested on the unseen test set to evaluate its generalization ability.\u003c/p\u003e \u003cp\u003eThe entire implementation was executed in R programming language within the RStudio Integrated Development Environment (IDE). The code used for data preprocessing, model training, and performance evaluation is included below.\u003c/p\u003e"},{"header":"III. RESULTS","content":"\u003cp\u003eThe performance of the trained model was evaluated using standard regression metrics, which provide a comprehensive assessment of its predictive accuracy. The results, presented in the table below, highlight the model\u0026rsquo;s ability to estimate retention time effectively based on the selected molecular descriptors.\u003c/p\u003e\u003cp\u003eTable 1. Performance metrics table\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS.No.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePerformance Metrics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean Squared Error\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e109.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eR-squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.609\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean Absolute Error\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e80.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e A scatter plot comparing actual vs. predicted retention times showed a dense clustering of points around the ideal prediction line. This distribution indicates that the model successfully captures the underlying relationship between retention time and molecular properties, demonstrating its reliability in making accurate predictions\u003c/p\u003e "},{"header":"IV. CONCLUSION","content":"\u003cp\u003eThis study demonstrates the effectiveness of Random Forest Regression in predicting HPLC retention time based on molecular descriptors. The model achieves an R\u0026sup2; value of 0.609, indicating a moderate predictive capability. Future work can improve accuracy by incorporating additional molecular descriptors and further optimizing hyperparameters, paving the way for integration of machine learning in chromatography analysis.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eV.G.S , A.R.U \u0026amp; T.C.B wrote and edited the main manuscript text. All authors reviewed the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe express our sincere gratitude to RV College of Engineering and RV Sikshana Samithi Trust.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe real time data set was obtained from : Kaggle, \u0026ldquo;Meltin retention times with molecular descriptors.\u0026rdquo; [Online]. Available: https://www.kaggle.com/datasets/satwikmurarka/meltin-retention-times-with-molecular-descriptors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eP. Patel, S. B. Narkhede, and S. Luhar, \u0026ldquo;High-performance liquid chromatography (HPLC): A comprehensive review,\u0026rdquo; EPRA Int. J. Res. Dev., pp. 85\u0026ndash;86, 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR. Ahmed, \u0026ldquo;High-Performance Liquid Chromatography (HPLC): Principles, applications, versatility, efficiency, innovation and comparative analysis in modern analytical chemistry and in pharmaceutical sciences,\u0026rdquo; 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH. A. Gavit, A. R. Pawar, V. V. Patil, P. S. Patil, and P. H. Sisodiya, \u0026ldquo;Advancement in High-Performance Liquid Chromatography Techniques: A comprehensive review,\u0026rdquo; Asian J. Pharm. Anal., vol. 95, pp. 95\u0026ndash;103, 2024, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.52711/2231-5675.2024.00017\u003c/span\u003e\u003cspan address=\"10.52711/2231-5675.2024.00017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026ldquo;Characterization of a high throughput approach for large scale retention measurement in liquid chromatography,\u0026rdquo; 2023, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.26434/chemrxiv-2023-czt1p-v3\u003c/span\u003e\u003cspan address=\"10.26434/chemrxiv-2023-czt1p-v3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026ldquo;RT-Tranformer: Retention time prediction for metabolite annotation to assist in metabolite identification,\u0026rdquo; 2023, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.26434/chemrxiv-2023-pf268\u003c/span\u003e\u003cspan address=\"10.26434/chemrxiv-2023-pf268\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ. L. E. Reubsaet and K. Jinno, \u0026ldquo;Characterisation of important interactions controlling retention behaviour of analytes in reversed-phase high-performance liquid chromatography,\u0026rdquo; TrAC Trends Anal. Chem., vol. 17, no. 3, pp. 157\u0026ndash;166, 1998, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S0165-9936(97)00127-1\u003c/span\u003e\u003cspan address=\"10.1016/S0165-9936(97)00127-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ. Golubović, A. Protić, B. Otašević, and M. Zečević, \u0026ldquo;Quantitative structure-retention relationships applied to development of liquid chromatography gradient-elution method for the separation of sartans,\u0026rdquo; Talanta, vol. 150, pp. 190\u0026ndash;197, 2016, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.talanta.2015.12.035\u003c/span\u003e\u003cspan address=\"10.1016/j.talanta.2015.12.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eV. David and A. Medvedovici, \u0026ldquo;Structure-retention correlation in liquid chromatography for pharmaceutical applications,\u0026rdquo; J. Liq. Chromatogr. Relat. Technol., vol. 30, pp. 761\u0026ndash;789, 2007, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/10826070701191052\u003c/span\u003e\u003cspan address=\"10.1080/10826070701191052\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eT. Bączek, R. Kaliszan, K. Novotn\u0026aacute;, and P. Jandera, \u0026ldquo;Comparative characteristics of HPLC columns based on quantitative structure-retention relationships (QSRR) and hydrophobic-subtraction model,\u0026rdquo; J. Chromatogr. A, vol. 1075, no. 1\u0026ndash;2, pp. 109\u0026ndash;115, 2005, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/J.CHROMA.2005.03.117\u003c/span\u003e\u003cspan address=\"10.1016/J.CHROMA.2005.03.117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eP. P. Basuri, K. Sahini, N. C. N, and G. V., \u0026ldquo;Retention behaviour of analytes in reversed phase high performance liquid chromatography - A review,\u0026rdquo; Biomed. Chromatogr., vol. 37, no. 7, p. e5482, 2022, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/bmc.5482\u003c/span\u003e\u003cspan address=\"10.1002/bmc.5482\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eY. Wang, \u003cem\u003eRetention Studies in HPLC\u003c/em\u003e, 1992.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eU. Judycka, K. Jagiello, M. Gromelski, L. Bober, J. Błażejowski, and T. Puzyn, \u0026ldquo;Chemometric outlook on correlations between retention parameters of polar and semipolar HPLC columns and physicochemical characteristics of ampholytic substances of biological and pharmaceutical relevance,\u0026rdquo; Struct. Chem., vol. 29, no. 6, pp. 1839\u0026ndash;1844, 2018, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/S11224-018-1174-5\u003c/span\u003e\u003cspan address=\"10.1007/S11224-018-1174-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS. Osipenko et al., \u0026ldquo;Machine learning to predict retention time of small molecules in nano-HPLC,\u0026rdquo; Anal. Bioanal. Chem., vol. 412, pp. 7767\u0026ndash;7776, 2020, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00216-020-02905-0\u003c/span\u003e\u003cspan address=\"10.1007/s00216-020-02905-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eL. Sun et al., \u0026ldquo;A simple method for HPLC retention time prediction: Linear calibration using two reference substances,\u0026rdquo; Chin. Med., vol. 12, 2017, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13020-017-0137-x\u003c/span\u003e\u003cspan address=\"10.1186/s13020-017-0137-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS. Osipenko, E. Nikolaev, and Y. Kostyukevich, \u0026ldquo;Retention time prediction with message-passing neural networks,\u0026rdquo; Separations, vol. 9, no. 10, p. 291, 2022, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/separations9100291\u003c/span\u003e\u003cspan address=\"10.3390/separations9100291\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM. Ros\u0026eacute;s, I. Canals, H. Allemann, K. Siigur, and E. Bosch, \u0026ldquo;Retention of ionizable compounds on HPLC. 2. Effect of pH, ionic strength, and mobile phase composition on the retention of weak acids,\u0026rdquo; Anal. Chem., vol. 68, no. 23, pp. 4094\u0026ndash;4100, 1996, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/ac960105d\u003c/span\u003e\u003cspan address=\"10.1021/ac960105d\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR. Ju et al., \u0026ldquo;Deep neural network pretrained by weighted autoencoders and transfer learning for retention time prediction of small molecules,\u0026rdquo; Anal. Chem., vol. 93, no. 47, pp. 15651\u0026ndash;15658, 2021, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/ACS.ANALCHEM.1C03250\u003c/span\u003e\u003cspan address=\"10.1021/ACS.ANALCHEM.1C03250\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eC. Ren et al., \u0026ldquo;Improved peptide retention time prediction in liquid chromatography through deep learning,\u0026rdquo; Anal. Chem., vol. 90, no. 18, pp. 10881\u0026ndash;10888, 2018, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/acs.analchem.8b02386\u003c/span\u003e\u003cspan address=\"10.1021/acs.analchem.8b02386\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR. Bouwmeester, L. Martens, and S. Degroeve, \u0026ldquo;Comprehensive and empirical evaluation of machine learning algorithms for small molecule LC retention time prediction,\u0026rdquo; Anal. Chem., vol. 91, no. 5, pp. 3694\u0026ndash;3703, 2019, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/acs.analchem.8b05820\u003c/span\u003e\u003cspan address=\"10.1021/acs.analchem.8b05820\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQ. Yang, H. Ji, H. Lu, and Z. Zhang, \u0026ldquo;Prediction of liquid chromatographic retention time with graph neural networks to assist in small molecule identification,\u0026rdquo; Anal. Chem., 2021, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1021/acs.analchem.0c04071\u003c/span\u003e\u003cspan address=\"10.1021/acs.analchem.0c04071\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQ. Yang, H. Ji, X. Fan, Z. Zhang, and H. Lu, \u0026ldquo;Retention time prediction in hydrophilic interaction liquid chromatography with graph neural network and transfer learning,\u0026rdquo; J. Chromatogr. A, vol. 1656, p. 462536, 2021, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.chroma.2021.462536\u003c/span\u003e\u003cspan address=\"10.1016/j.chroma.2021.462536\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeeksforGeeks, \u0026ldquo;Random forest regression in Python.\u0026rdquo; [Online]. Available: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.geeksforgeeks.org/random-forest-regression-in-python/\u003c/span\u003e\u003cspan address=\"https://www.geeksforgeeks.org/random-forest-regression-in-python/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eN. Goudarzi, D. Shahsavani, F. Emadi-Gandaghi, and M. Chamjangali, \u0026ldquo;Quantitative structure-property relationships of retention indices of some sulfur organic compounds using random forest technique as a variable selection and modeling method,\u0026rdquo; J. Sep. Sci., vol. 39, no. 19, pp. 3835\u0026ndash;3842, 2016, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/jssc.201600358\u003c/span\u003e\u003cspan address=\"10.1002/jssc.201600358\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eN. Goudarzi, D. Shahsavani, F. Emadi-Gandaghi, and M. Chamjangali, \u0026ldquo;Application of random forests method to predict the retention indices of some polycyclic aromatic hydrocarbons,\u0026rdquo; J. Chromatogr. A, vol. 1333, pp. 25\u0026ndash;31, 2014, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.chroma.2014.01.048\u003c/span\u003e\u003cspan address=\"10.1016/j.chroma.2014.01.048\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eN. Goudarzi and D. Shahsavani, \u0026ldquo;Application of a random forests (RF) method as a new approach for variable selection and modelling in a QSRR study to predict the relative retention time of some polybrominated diphenylethers (PBDEs),\u0026rdquo; Anal. Methods, vol. 4, pp. 3733\u0026ndash;3738, 2012, doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1039/C2AY25484K\u003c/span\u003e\u003cspan address=\"10.1039/C2AY25484K\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM. Kursa, L. Komsta, and W. Rudnicki, \u0026ldquo;Random forest models of the retention constants in thin layer chromatography,\u0026rdquo; arXiv preprint, arXiv:1106.3361, 2011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eE. Bandini et al., \u0026ldquo;Physicochemical modelling of retention mechanisms using machine learning,\u0026rdquo; J. Cheminform., vol. 16, p. 72, 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaggle, \u0026ldquo;Meltin retention times with molecular descriptors.\u0026rdquo; [Online]. Available: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.kaggle.com/datasets/satwikmurarka/meltin-retention-times-with-molecular-descriptors\u003c/span\u003e\u003cspan address=\"https://www.kaggle.com/datasets/satwikmurarka/meltin-retention-times-with-molecular-descriptors\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"High-Performance Liquid Chromatography, retention time, Random Forest Regression, molecular descriptors","lastPublishedDoi":"10.21203/rs.3.rs-6053718/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6053718/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHigh-Performance Liquid Chromatography is widely used for compound analysis, where retention time (RT) serves as a critical parameter. This study employs a Random Forest Regression model to predict RT based on molecular descriptors such as molecular weight, partial charge, partition coefficient, and topological polar surface area. The model successfully predicted the retention time with high similarity to the real data, thereby validating its accuracy. This study highlights the potential of machine learning in optimizing chromatographic analysis.\u003c/p\u003e","manuscriptTitle":"Retention Time Prediction in High-Performance Liquid Chromatography Using Random Forest Regression","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-05 07:10:42","doi":"10.21203/rs.3.rs-6053718/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bf1a0bf2-d63c-4ec0-99fc-37a7ecc0f456","owner":[],"postedDate":"March 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-03-05T07:10:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-05 07:10:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6053718","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6053718","identity":"rs-6053718","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}